Abstract: This paper proposes a novel confusion and diffusion method for image encryption. One innovation is to confuse the pixels by transforming the nucleotide into its base pair for random times, the other is to generate the new keys according to the plain image and the common keys, which can make the initial conditions of the chaotic maps change automatically in every encryption process. For any size of the original grayscale image, after being permuted the rows and columns respectively by the arrays generated by piecewise linear chaotic map (PWLCM), each pixel of the original image is encoded into four nucleotides by the deoxyribonucleic acid (DNA) coding, then each nucleotide is transformed into its base pair for random time(s) using the complementary rule, the times is generated by Chebyshev maps. Experiment results and security analysis show that the scheme can not only achieve good encryption result, but also the key space is large enough to resist against common attacks.

Abstract: In these days, considering the growth of knowledge about sustainability in enterprise, the sustainable supplier selection would be the central component in the management of a sustainable supply chain. In this paper the sustainable supplier selection criteria and sub-criteria are determined and based on those criteria and sub-criteria a methodology is proposed onto evaluation and ranking of a given set of suppliers. In the evaluation process, decision makers' opinions on the importance of deciding the criteria and sub-criteria, in addition to their preference of the suppliers' performance with respect to sub-criteria are considered in linguistic terms. To handle the subjectivity of decision makers' assessments, fuzzy logic has been applied and a new ranking method on the basis of fuzzy inference system (FIS) is proposed for supplier selection problem. Finally, an illustrative example is utilized to show the feasibility of the proposed method.

Abstract: A common algorithm to solve the shortest path problem (SPP) is the Dijkstra algorithm. In this paper, a generalized Dijkstra algorithm is proposed to handle SPP in an uncertain environment. Two key issues need to be addressed in SPP with fuzzy parameters. One is how to determine the addition of two edges. The other is how to compare the distance between two different paths with their edge lengths represented by fuzzy numbers. To solve these problems, the graded mean integration representation of fuzzy numbers is adopted to improve the classical Dijkstra algorithm. A numerical example of a transportation network is used to illustrate the efficiency of the proposed method.

Abstract: The engineering properties of the rocks have the most vital role in planning of rock excavation and construction for optimum utilization of earth resources with greater safety and least damage to surroundings. The design and construction of structure is influenced by physico-mechanical properties of rock mass. Young's modulus provides insight about the magnitude and characteristic of the rock mass deformation due to change in stress field. The determination of the Young's modulus in laboratory is very time consuming and costly. Therefore, basic rock properties like point load, density and water absorption have been used to predict the Young's modulus. Point load, density and water absorption can be easily determined in field as well as laboratory and are pertinent properties to characterize a rock mass. The artificial neural network (ANN), fuzzy inference system (FIS) and neuro fuzzy are promising techniques which have proven to be very reliable in recent years. In, present study, neuro fuzzy system is applied to predict the rock Young's modulus to overcome the limitation of ANN and fuzzy logic. Total 85 dataset were used for training the network and 10 dataset for testing and validation of network rules. The network performance indices correlation coefficient, mean absolute percentage error (MAPE), root mean square error (RMSE), and variance account for (VAF) are found to be 0.6643, 7.583, 6.799, and 91.95 respectively, which endow with high performance of predictive neuro-fuzzy system to make use for prediction of complex rock parameter.

Abstract: This paper deals with the application of artificial neural network (ANN) based ANFIS approach to automatic generation control (AGC) of a three unequal area hydrothermal system. The proposed ANFIS controller combines the advantages of fuzzy controller as well as quick response and adaptability nature of ANN. Appropriate generation rate constraints (GRC) have been considered for the thermal and hydro plants. The hydro area is considered with an electric governor and thermal area is considered with reheat turbine. The design objective is to improve the frequency and tie-line power deviations of the interconnected system. 1% step load perturbation has been considered occurring either in any individual area or occurring simultaneously in all the areas. It is a maiden application of ANFIS approach to a three unequal area hydrothermal system with GRC considering perturbation in a single area as well as in all areas. The performance of the ANFIS controller is compared with the results of integral squared error (ISE) criterion based integral controller published previously. Simulation results are presented to show the improved performance of ANFIS controller in comparison with the conventional integral controller. The results indicate that the controllers exhibit better performance. In fact, ANFIS approach satisfies the load frequency control requirements with a reasonable dynamic response.

Abstract: We give the explicit solutions of uncertain fractional differential equations (UFDEs) under Riemann–Liouville H-differentiability using Mittag-Leffler functions. To this end, Riemann–Liouville H-differentiability is introduced which is a direct generalization of the concept of Riemann–Liouville differentiability in deterministic sense to the fuzzy context. Moreover, equivalent integral forms of UFDEs are determined which are applied to derive the explicit solutions. Finally, some illustrative examples are given.

Abstract: In most of the industries related to mechanical engineering, the usage of pumps is high. Hence, the system which takes care of the continuous running of the pump becomes essential. In this paper, a vibration based condition monitoring system is presented for monoblock centrifugal pumps as it plays relatively critical role in most of the industries. This approach has mainly three steps namely feature extraction, classification and comparison of classification. In spite of availability of different efficient algorithms for fault detection, the wavelet analysis for feature extraction and Naive Bayes algorithm and Bayes net algorithm for classification is taken and compared. This paper presents the use of Naive Bayes algorithm and Bayes net algorithm for fault diagnosis through discrete wavelet features extracted from vibration signals of good and faulty conditions of the components of centrifugal pump. The classification accuracies of different discrete wavelet families were calculated and compared to find the best wavelet for the fault diagnosis of the centrifugal pump.

Abstract: In the areas of investment research and applications, feasible quantitative models include methodologies stemming from soft computing for prediction of financial time series, multi-objective optimization of investment return and risk reduction, as well as selection of investment instruments for portfolio management based on asset ranking using a variety of input variables and historical data, etc. Among all these, stock selection has long been identified as a challenging and important task. This line of research is highly contingent upon reliable stock ranking for successful portfolio construction. Recent advances in machine learning and data mining are leading to significant opportunities to solve these problems more effectively. In this study, we aim at developing a methodology for effective stock selection using support vector regression (SVR) as well as genetic algorithms (GAs). We first employ the SVR method to generate surrogates for actual stock returns that in turn serve to provide reliable rankings of stocks. Top-ranked stocks can thus be selected to form a portfolio. On top of this model, the GA is employed for the optimization of model parameters, and feature selection to acquire optimal subsets of input variables to the SVR model. We will show that the investment returns provided by our proposed methodology significantly outperform the benchmark. Based upon these promising results, we expect this hybrid GA-SVR methodology to advance the research in soft computing for finance and provide an effective solution to stock selection in practice.

Abstract: Intrusion detection system (IDS) is to monitor the attacks occurring in the computer or networks. Anomaly intrusion detection plays an important role in IDS to detect new attacks by detecting any deviation from the normal profile. In this paper, an intelligent algorithm with feature selection and decision rules applied to anomaly intrusion detection is proposed. The key idea is to take the advantage of support vector machine (SVM), decision tree (DT), and simulated annealing (SA). In the proposed algorithm, SVM and SA can find the best selected features to elevate the accuracy of anomaly intrusion detection. By analyzing the information from using KDD'99 dataset, DT and SA can obtain decision rules for new attacks and can improve accuracy of classification. In addition, the best parameter settings for the DT and SVM are automatically adjusted by SA. The proposed algorithm outperforms other existing approaches. Simulation results demonstrate that the proposed algorithm is successful in detecting anomaly intrusion detection.

Abstract: The motor unit action potentials (MUPs) in an electromyographic (EMG) signal provide a significant source of information for the assessment of neuromuscular disorders. Since recently there were different types of developments in computer-aided EMG equipment, different methodologies in the time domain and frequency domain has been followed for quantitative analysis of EMG signals. In this study, the usefulness of the different feature extraction methods for describing MUP morphology is investigated. Besides, soft computing techniques were presented for the classification of intramuscular EMG signals. The proposed method automatically classifies the EMG signals into normal, neurogenic or myopathic. Also, multilayer perceptron neural networks (MLPNN), dynamic fuzzy neural network (DFNN) and adaptive neuro-fuzzy inference system (ANFIS) based classifiers were compared in relation to their accuracy in the classification of EMG signals. Concerning the impacts of features on the EMG signal classification, different results were obtained through analysis of the soft computing techniques. The comparative analysis suggests that the ANFIS modelling is superior to the DFNN and MLPNN in at least three points: slightly higher recognition rate; insensitivity to overtraining; and consistent outputs demonstrating higher reliability.

Abstract: Artificial bee colony (ABC) algorithm is one of the recently proposed swarm intelligence based algorithms for continuous optimization. Therefore it is not possible to use the original ABC algorithm directly to optimize binary structured problems. In this paper we introduce a new version of ABC, called DisABC, which is particularly designed for binary optimization. DisABC uses a new differential expression, which employs a measure of dissimilarity between binary vectors in place of the vector subtraction operator typically used in the original ABC algorithm. Such an expression helps to maintain the major characteristics of the original one and is respondent to the structure of binary optimization problems, too. Similar to original ABC algorithm, DisABC's differential expression works in continuous space while its consequence is used in a two-phase heuristic to construct a complete solution in binary space. Effectiveness of DisABC algorithm is tested on solving the uncapacitated facility location problem (UFLP). A set of 15 benchmark test problem instances of UFLP are adopted from OR-Library and solved by the proposed algorithm. Results are compared with two other state of the art binary optimization algorithms, i.e., binDE and PSO algorithms, in terms of three quality indices. Comparisons indicate that DisABC performs very well and can be regarded as a promising method for solving wide class of binary optimization problems.

Abstract: This study intends to use a combination of fuzzy Analytic Hierarchy Process (AHP) and fuzzy Decision-making Trial and Evaluation Laboratory (DEMATEL) method in human resource for science and technology (HRST). Specifically, this study first uses AHP to evaluate the weighting for each criterion and then use DEMATEL method to establish contextual relationships among those criteria. We find out Infrastructure might be more critical since it is a cause and will directly influence human resource for science and technology performance. For human resource for science and technology (HRST), improving Infrastructure might be a better choice for the long period of time. Moreover, Education, R&D Expenses and Immediate output are more important second-tier criteria than Value, Cooperation, Labor Market, Human Capital and Intermediate output. Therefore, the improvement should be started with Infrastructure, particularly on identification of the Education, R&D Expenses and Immediate output.

Abstract: Artificial bee colony (ABC) algorithm is an optimization algorithm based on the intelligent foraging behavior of honeybee swarm. Optimal reactive power flow (ORPF) based on ABC algorithm to minimize active power loss in power systems is studied in this paper. The advantage of ABC algorithm is that it does not require these parameters, because it is very difficult to determine external parameters such as cross over rate and mutation rate as in case of genetic algorithm and differential evolution. The other advantage is that global search ability of the algorithm is implemented by introducing a neighborhood source production mechanism which is similar to mutation process. Because of these features, ABC algorithm attracts much attention in recent years and has been used successfully in many areas. ORPF problem is one of these areas. In this paper, proposed algorithm is tested on both standard IEEE 30-bus test system and IEEE 118-bus test system. To show the effectiveness of proposed algorithms, the obtained results are compared with different approaches as available in the literature.

Abstract: Identification of local regions from where optimal discriminating features can be extracted is one of the major tasks in the area of pattern recognition. To locate such regions different kind of region sampling techniques are used in the literature. There is no standard methodology to identify exactly such regions. Here we have proposed a methodology where local regions of varying heights and widths are created dynamically. Genetic algorithm (GA) is then applied on these local regions to sample the optimal set of local regions from where an optimal feature set can be extracted that has the best discriminating features. We have evaluated the proposed methodology on a data set of handwritten Bangla digits. In the present work, we have randomly generated seven sets of local regions and from every set, GA selects an optimal group of local regions which produces best recognition performance with a support vector machine (SVM) based classifier. Other popular optimization techniques like simulated annealing (SA) and hill climbing (HC) have also been evaluated with the same data set and maximum recognition accuracies were found to be 97%, 96.7% and 96.7% for GA, SA and HC, respectively. We have also compared the performance of the present technique with those of other zone based techniques on the same database.

Abstract: This paper performs an unprecedented comparative study of Metaheuristic optimization frameworks. As criteria for comparison a set of 271 features grouped in 30 characteristics and 6 areas has been selected. These features include the different metaheuristic techniques covered, mechanisms for solution encoding, constraint handling, neighborhood specification, hybridization, parallel and distributed computation, software engineering best practices, documentation and user interface, etc. A metric has been defined for each feature so that the scores obtained by a framework are averaged within each group of features, leading to a final average score for each framework. Out of 33 frameworks ten have been selected from the literature using well-defined filtering criteria, and the results of the comparison are analyzed with the aim of identifying improvement areas and gaps in specific frameworks and the whole set. Generally speaking, a significant lack of support has been found for hyper-heuristics, and parallel and distributed computing capabilities. It is also desirable to have a wider implementation of some Software Engineering best practices. Finally, a wider support for some metaheuristics and hybridization capabilities is needed.

Abstract: Knowledge is a key source of sustainable competitive advantage. In response to increasingly drastic and competitive environments, many organizations wish to better utilize and manage knowledge for business success. For the purpose to execute formal knowledge management (KM) effectively, some works have suggested several critical factors of KM implementations. However, in a strategic view, such a list of critical factors must be further honed to increase practical usefulness, as not all critical factors necessarily share the same importance. Moreover, assessing the importance of critical factors inevitably involves the vagueness of human judgment. Hence, this study presents a favorable method combining fuzzy set theory and the Decision Making Trial and Evaluation Laboratory (DEMATEL) method to segment the critical factors for successful KM implementations. Also, an empirical study is presented to illustrate the proposed method and to demonstrate its usefulness.

Abstract: A regularity model-based multiobjective estimation of distribution algorithm (RM-MEDA) has been proposed for solving continuous multiobjective optimization problems with variable linkages. RM-MEDA is a kind of estimation of distribution algorithms and, therefore, modeling plays a critical role. In RM-MEDA, the population is split into several clusters to build the model. Moreover, the fixed number of clusters is recommended in RM-MEDA when solving different kinds of problems. However, based on our experiments, we find that the number of clusters is problem-dependent and has a significant effect on the performance of RM-MEDA. Motivated by the above observation, in this paper we improve the clustering process and propose a reducing redundant cluster operator (RRCO) to build more precise model during the evolution. By combining RRCO with RM-MEDA, we present an improved version of RM-MEDA, named IRM-MEDA. In this paper, we also construct four additional continuous multiobjective optimization test instances. The experimental results have shown that IRM-MEDA outperforms RM-MEDA in terms of efficiency and effectiveness. In particular, IRM-MEDA performs on average 31.67% faster than RM-MEDA.

Abstract: This paper presents a novel two-stage hybrid swarm intelligence optimization algorithm called GA---PSO---ACO algorithm that combines the evolution ideas of the genetic algorithms, particle swarm optimization and ant colony optimization based on the compensation for solving the traveling salesman problem. In the proposed hybrid algorithm, the whole process is divided into two stages. In the first stage, we make use of the randomicity, rapidity and wholeness of the genetic algorithms and particle swarm optimization to obtain a series of sub-optimal solutions (rough searching) to adjust the initial allocation of pheromone in the ACO. In the second stage, we make use of these advantages of the parallel, positive feedback and high accuracy of solution to implement solving of whole problem (detailed searching). To verify the effectiveness and efficiency of the proposed hybrid algorithm, various scale benchmark problems from TSPLIB are tested to demonstrate the potential of the proposed two-stage hybrid swarm intelligence optimization algorithm. The simulation examples demonstrate that the GA---PSO---ACO algorithm can greatly improve the computing efficiency for solving the TSP and outperforms the Tabu Search, genetic algorithms, particle swarm optimization, ant colony optimization, PS---ACO and other methods in solution quality. And the experimental results demonstrate that convergence is faster and better when the scale of TSP increases.

Abstract: Stochastic optimization algorithms like genetic algorithms (GAs) and particle swarm optimization (PSO) algorithms perform global optimization but waste computational effort by doing a random search. On the other hand deterministic algorithms like gradient descent converge rapidly but may get stuck in local minima of multimodal functions. Thus, an approach that combines the strengths of stochastic and deterministic optimization schemes but avoids their weaknesses is of interest. This paper presents a new hybrid optimization algorithm that combines the PSO algorithm and gradient-based local search algorithms to achieve faster convergence and better accuracy of final solution without getting trapped in local minima. In the new gradient-based PSO algorithm, referred to as the GPSO algorithm, the PSO algorithm is used for global exploration and a gradient based scheme is used for accurate local exploration. The global minimum is located by a process of finding progressively better local minima. The GPSO algorithm avoids the use of inertial weights and constriction coefficients which can cause the PSO algorithm to converge to a local minimum if improperly chosen. The De Jong test suite of benchmark optimization problems was used to test the new algorithm and facilitate comparison with the classical PSO algorithm. The GPSO algorithm is compared to four different refinements of the PSO algorithm from the literature and shown to converge faster to a significantly more accurate final solution for a variety of benchmark test functions.

Abstract: Drought forecasts can be an effective tool for mitigating some of the more adverse consequences of drought. Data-driven models are suitable forecasting tools due to their rapid development times, as well as minimal information requirements compared to the information required for physically based models. This study compares the effectiveness of three data-driven models for forecasting drought conditions in the Awash River Basin of Ethiopia. The Standard Precipitation Index (SPI) is forecast and compared using artificial neural networks (ANNs), support vector regression (SVR), and wavelet neural networks (WN). SPI 3 and SPI 12 were the SPI values that were forecasted. These SPI values were forecast over lead times of 1 and 6 months. The performance of all the models was compared using RMSE, MAE, and R2. The forecast results indicate that the coupled wavelet neural network (WN) models were the best models for forecasting SPI values over multiple lead times in the Awash River Basin in Ethiopia.

Abstract: Economic dispatch is carried out at the energy control center to find out the optimal output of thermal generating units such that power balance criterion is met, unit operating limits are satisfied and the fuel cost is minimized. With growing environmental awareness and strict government regulations throughout the world, it has become essential to optimize not only the total fuel cost but also the harmful emissions, both, under static as well as dynamic conditions. The static environment economic dispatch finds the optimal output of generating units for a fixed load demand at a given time, while the dynamic environmental economic dispatch schedules the output of online generators with changing power demands over a certain time period (normally one day) so as to minimize these two conflicting objectives, simultaneously. In this paper, the price penalty factor approach is employed for simultaneous minimization of cost and emission. The generator ramp rate constraints, non-convex and discontinuous nature of cost function and the large number of generators in practical power plants, make this problem very difficult to solve. Here, a fuzzy ranking approach is employed to identify the solution which offers the best compromise between cost and emission objectives. This paper proposes an improved bacterial foraging algorithm (IBFA) in which a parameter automation strategy and crossover operation is used in micro BFA to improve computational efficiency. The performance of IBFA is compared with classical BFA and with previously published literature on four standard test systems and is found to be better.

Abstract: A mobile ad hoc network (MANET) is dynamic in nature and is composed of wirelessly connected nodes that perform hop-by-hop routing without the help of any fixed infrastructure. One of the important requirements of a MANET is the efficiency of energy, which increases the lifetime of the network. Several techniques have been proposed by researchers to achieve this goal and one of them is clustering in MANETs that can help in providing an energy-efficient solution. Clustering involves the selection of cluster-heads (CHs) for each cluster and fewer CHs result in greater energy efficiency as these nodes drain more power than noncluster-heads. In the literature, several techniques are available for clustering by using optimization and evolutionary techniques that provide a single solution at a time. In this paper, we propose a multi-objective solution by using multi-objective particle swarm optimization (MOPSO) algorithm to optimize the number of clusters in an ad hoc network as well as energy dissipation in nodes in order to provide an energy-efficient solution and reduce the network traffic. In the proposed solution, inter-cluster and intra-cluster traffic is managed by the cluster-heads. The proposed algorithm takes into consideration the degree of nodes, transmission power, and battery power consumption of the mobile nodes. The main advantage of this method is that it provides a set of solutions at a time. These solutions are achieved through optimal Pareto front. We compare the results of the proposed approach with two other well-known clustering techniques; WCA and CLPSO-based clustering by using different performance metrics. We perform extensive simulations to show that the proposed approach is an effective approach for clustering in mobile ad hoc networks environment and performs better than the other two approaches.

Abstract: In this paper, we present the induced generalized intuitionistic fuzzy ordered weighted averaging (I-GIFOWA) operator. It is a new aggregation operator that generalized the IFOWA operator, including all the characteristics of both the generalized IFOWA and the induced IFOWA operators. It provides a very general formulation that includes as special cases a wide range of aggregation operators for intuitionistic fuzzy information, including all the particular cases of the I-IFOWA operator, GIFOWA operator and the induced intuitionistic fuzzy ordered geometric (I-IFOWG) operator. We also present the induced generalized interval-valued intuitionistic fuzzy ordered weighted averaging (I-GIIFOWA) operator to accommodate the environment in which the given arguments are interval-valued intuitionistic fuzzy sets. Further, we develop procedures to apply them to solve group multiple attribute decision making problems with intuitionistic fuzzy or interval-valued intuitionistic fuzzy information. Finally, we present their application to show the effectiveness of the developed methods.

Abstract: This paper presents a new approach to classify fault types and predict the fault location in the high-voltage power transmission lines, by using Support Vector Machines (SVM) and Wavelet Transform (WT) of the measured one-terminal voltage and current transient signals. Wavelet entropy criterion is applied to wavelet detail coefficients to reduce the size of feature vector before classification and prediction stages. The experiments performed for different kinds of faults occurred on the transmission line have proved very good accuracy of the proposed fault location algorithm. The fault classification error is below 1% for all tested fault conditions. The average error of fault location in a 380kV-360-km transmission line is below 0.26% and the maximum error did not exceed 0.95km.

Abstract: Combinatorial optimization problems are usually modeled in a static fashion. In this kind of problems, all data are known in advance, i.e. before the optimization process has started. However, in practice, many problems are dynamic, and change while the optimization is in progress. For example, in the dynamic vehicle routing problem (DVRP), new orders arrive when the working day plan is in progress. In this case, routes must be reconfigured dynamically while executing the current simulation. The DVRP is an extension of a conventional routing problem, its main interest being the connection to many real word applications (repair services, courier mail services, dial-a-ride services, etc.). In this article, a DVRP is examined, and solving methods based on particle swarm optimization and variable neighborhood search paradigms are proposed. The performance of both approaches is evaluated using a new set of benchmarks that we introduce here as well as existing benchmarks in the literature. Finally, we measure the behavior of both methods in terms of dynamic adaptation.

Abstract: Recommender systems, which recommend items of information that are likely to be of interest to the users, and filter out less favored data items, have been developed. Collaborative filtering is a widely used recommendation technique. It is based on the assumption that people who share the same preferences on some items tend to share the same preferences on other items. Clustering techniques are commonly used for collaborative filtering recommendation. While cluster ensembles have been shown to outperform many single clustering techniques in the literature, the performance of cluster ensembles for recommendation has not been fully examined. Thus, the aim of this paper is to assess the applicability of cluster ensembles to collaborative filtering recommendation. In particular, two well-known clustering techniques (self-organizing maps (SOM) and k-means), and three ensemble methods (the cluster-based similarity partitioning algorithm (CSPA), hypergraph partitioning algorithm (HGPA), and majority voting) are used. The experimental results based on the Movielens dataset show that cluster ensembles can provide better recommendation performance than single clustering techniques in terms of recommendation accuracy and precision. In addition, there are no statistically significant differences between either the three SOM ensembles or the three k-means ensembles. Either the SOM or k-means ensembles could be considered in the future as the baseline collaborative filtering technique.

Abstract: The necessity of sewers to carry sediment has been recognized for many years. Typically, old sewage systems were designated based on self-cleansing concept where there is no deposition in sewer. These codes were applicable to non-cohesive sediments (typically storm sewers). This study presents adaptive neuro-fuzzy inference system (ANFIS), which is a combination of neural network and fuzzy logic, as an alternative approach to predict the functional relationships of sediment transport in sewer pipe systems. The proposed relationship can be applied to different boundaries with partially full flow. The present ANFIS approach gives satisfactory results (r^2=0.98 and RMSE=0.002431) compared to the existing predictor.

Abstract: Financial distress prediction (FDP) is of great importance to both inner and outside parts of companies. Though lots of literatures have given comprehensive analysis on single classifier FDP method, ensemble method for FDP just emerged in recent years and needs to be further studied. Support vector machine (SVM) shows promising performance in FDP when compared with other single classifier methods. The contribution of this paper is to propose a new FDP method based on SVM ensemble, whose candidate single classifiers are trained by SVM algorithms with different kernel functions on different feature subsets of one initial dataset. SVM kernels such as linear, polynomial, RBF and sigmoid, and the filter feature selection/extraction methods of stepwise multi discriminant analysis (MDA), stepwise logistic regression (logit), and principal component analysis (PCA) are applied. The algorithm for selecting SVM ensemble's base classifiers from candidate ones is designed by considering both individual performance and diversity analysis. Weighted majority voting based on base classifiers' cross validation accuracy on training dataset is used as the combination mechanism. Experimental results indicate that SVM ensemble is significantly superior to individual SVM classifier when the number of base classifiers in SVM ensemble is properly set. Besides, it also shows that RBF SVM based on features selected by stepwise MDA is a good choice for FDP when individual SVM classifier is applied.

Abstract: This study presented a new diagnosis system for myocardial infarction classification by converting multi-lead ECG data into a density model for increasing accuracy and flexibility of diseases detection. In contrast to the traditional approaches, a hybrid system with HMMs and GMMs was employed for data classification. A hybrid approach using multi-leads, i.e., lead-V1, V2, V3 and V4 for myocardial infarction were developed and HMMs were used not only to find the ECG segmentations but also to calculate the log-likelihood value which was treated as statistical feature data of each heartbeat's ECG complex. The 4-dimension feature vector extracted by HMMs was clustered by GMMs with different numbers of distribution (disease and normal data). SVMs classifier was also examined for comparison with our system in experimental result. There were total 1129 samples of heartbeats from clinical data, including 582 data with myocardial infarction and 547 normal data. The sensitivity of this diagnosis system achieved 85.71%, specificity achieved 79.82% and accuracy achieved 82.50% statistically.